Ebola Virus Deep Dive: Virology, Pathogenesis, Outbreaks and the Evidence for Vaccines and Treatments

Disclaimer: This guide is general scientific information only and is not medical advice. Clinical decisions about suspected Ebola disease must be made by qualified health professionals following WHO and national guidance, with laboratory confirmation.

Read the companion guide: For the plain-language version — what Ebola is, how it spreads, symptoms, vaccines and myths corrected — see our public guide: Ebola Virus: Symptoms, How It Spreads, Vaccines and Treatment — A Complete Guide

Taxonomy: one family name, several different viruses

Ebolaviruses belong to the family Filoviridae (filament-shaped viruses), order Mononegavirales. In the International Committee on Taxonomy of Viruses (ICTV) classification updated in 2023, the genus formerly called Ebolavirus was renamed Orthoebolavirus, and its species were given binomial names. The genus contains six species, each with a single member virus:

The older names — Zaire, Sudan and Bundibugyo ebolavirus — remain common in the literature and drug labels, so readers will meet both systems. The distinction is not pedantry: it determines which countermeasures apply. Licensed vaccines and approved antibodies target Ebola virus (EBOV) specifically; WHO is explicit that approved vaccines and treatments exist only for Ebola virus, with Sudan and Bundibugyo products still in development. Much public confusion about the 2026 outbreak comes from ignoring this species boundary.

Virology: a filamentous RNA virus built around one surface protein

Ebola virus is a negative-sense, single-stranded RNA virus — its genome is a single RNA molecule of approximately 19 kilobases, read in reverse to make messenger RNA. The genome encodes seven principal proteins, in order: nucleoprotein (NP), polymerase cofactor VP35, matrix protein VP40, surface glycoprotein (GP), transcription activator VP30, interferon antagonist VP24, and the RNA-dependent RNA polymerase (L). The virion itself is filamentous — long, thread-like particles, sometimes branched or bent — with a lipid envelope taken from the host cell membrane, studded with GP spikes. Filovirus particles are handled at biosafety level 4, the highest containment level, and Ebola virus is classified as a WHO Risk Group 4 pathogen.

GP is the protein that matters most, clinically and scientifically. It is the only viral protein on the virion surface, it mediates cell entry, it is the target of the licensed vaccine and of both approved antibody treatments, and it is the protein whose sequence differences between species limit cross-protection. GP is produced as a precursor that is cleaved into two subunits, GP1 (which handles attachment) and GP2 (which drives membrane fusion). Through RNA editing of the GP gene, infected cells also produce a secreted, soluble form of the glycoprotein (sGP), which is released in large amounts into the bloodstream and is thought to act partly as a decoy, absorbing antibodies that might otherwise neutralise the virus. GP also interacts with innate immune signalling, including Toll-like receptor 4 (TLR4)-dependent inflammatory pathways in macrophages, which may contribute to the dysregulated inflammation seen in severe disease.

Inside the cell, the genome is wrapped by NP into a helical nucleocapsid with VP35, VP30, VP24 and the L polymerase. Transcription and replication occur in cytoplasmic inclusion bodies. VP30 enables efficient transcription, VP35 is a polymerase cofactor and immune antagonist, and L carries out RNA synthesis. New virions assemble at the cell membrane, where VP40 drives budding.

Cell entry: a multi-step route through the endosome

Ebola virus entry is unusually indirect, and each step is a potential drug target that has so far resisted translation into a licensed medicine. The virus first attaches to the cell surface through several attachment factors, including phosphatidylserine receptors such as TIM-1 (HAVCR1) and C-type lectins on macrophages and dendritic cells, and is taken up by a macropinocytosis-like process — the cell essentially drinks in the particle. Inside the acidified endosome, host proteases called cathepsins B and L cleave GP1, stripping away its glycan cap and mucin-like domain and exposing the receptor-binding site. Only then can cleaved GP bind its essential intracellular receptor, the late endosomal cholesterol transporter Niemann–Pick C1 (NPC1). Binding to NPC1 triggers GP2-mediated fusion of the viral envelope with the endosomal membrane, releasing the nucleocapsid into the cytoplasm.

NPC1 is broadly expressed across cell types, which helps explain the virus’s wide tissue tropism once infection is established, although receptor expression alone does not determine which cells support sustained replication or long-term persistence. Genome-wide screens have also identified additional host factors for entry and internalisation, including the transporter SLC39A9 and the kinase PIK3C3, whose activity appears indispensable for internalisation of Ebola virions. None of these host-directed entry mechanisms has yet produced an approved Ebola drug, and that gap matters: every licensed countermeasure instead targets GP from outside the cell.

Pathogenesis: how Ebola defeats the immune system and kills

Ebola virus’s first targets are the very cells meant to detect it: monocytes, macrophages and dendritic cells. Infected macrophages and dendritic cells become factories for new virus while failing to sound the alarm, because two viral proteins systematically disable the interferon system — the body’s early antiviral warning network:

• VP35 binds double-stranded RNA, the molecular pattern that signals “virus present”, and masks it from the cellular sensors RIG-I and MDA5 that would otherwise trigger interferon production. It also functions as a polymerase cofactor, so the same protein that helps the virus copy itself also hides it.
• VP24 blocks the downstream response: it prevents phosphorylated STAT1 from being transported into the nucleus, so even interferon that is produced cannot switch on antiviral genes in the infected cell. VP24 also contributes to nucleocapsid assembly.

Dendritic cells, which should mature and present viral antigens to T cells, remain incompletely activated, so adaptive immunity is delayed and defective. The result is unchecked early replication and very high viral loads — and viral load at presentation is one of the strongest predictors of death.

As infection spreads to the liver, spleen, lymph nodes and endothelium, silent replication gives way to dysregulated inflammation: infected macrophages release large amounts of inflammatory mediators, endothelial function is disrupted, coagulation is disturbed (producing both clotting abnormalities and bleeding), and fluid leaks from blood vessels. Patients develop profound gastrointestinal fluid losses, electrolyte disturbance, falling blood pressure, kidney and liver failure, encephalopathy and, in some, haemorrhage. Death most often follows hypovolaemic shock and multi-organ failure rather than exsanguination. Fatal cases typically show higher viral loads, earlier severe symptoms, and higher creatinine and liver enzyme levels at presentation — all markers that also predicted death in the PALM trial.

Recovery is not always the end of the story. Ebola virus can persist in immune-privileged sites — the testes, the inside of the eye, the central nervous system, and in pregnancy, the placenta, amniotic fluid and fetus — after clearance from the blood. Viral RNA has been detected in semen more than two years after onset in some survivors, and sexual transmission has been documented up to fifteen months after clinical recovery. Relapse — symptomatic illness without re-infection, apparently from a persistent reservoir — is rare but documented, and the unusually low genetic divergence of the 2021 Guinea outbreak lineage from viruses of the 2013–2016 epidemic is consistent with a prolonged period of persistence before resurgence. These observations have changed how Ebola is understood: it is not only an acute infection but, in a minority of survivors, a virus with a long tail that requires structured survivor care, semen testing programmes and long-term surveillance.

Epidemiology: the major outbreaks

The table below covers the outbreaks that define Ebola’s history. Case fatality rates (CFR) are crude ratios of deaths to reported cases and are not directly comparable across outbreaks: they reflect the virus species, the speed of detection, access to care, and how completely mild cases were counted.

Three patterns stand out. First, Ebola virus (Zaire) outbreaks have generally been the largest, but the 2026 Bundibugyo outbreak broke that pattern and is now the second-largest Ebola outbreak of any species on record, after West Africa. Second, fatality rates have not fallen steadily over time: the 2025 Kasai outbreak (70.3%) and the 2018–2020 Kivu outbreak (66%) were deadlier, proportionally, than West Africa (~40%), a reminder that CFR tracks context — late presentation, conflict, health-system capacity — as much as biology. In the 2026 outbreak, WHO has repeatedly noted that a large share of deaths occur in communities rather than treatment centres (307 of 482 deaths in one October 2026 reporting interval), pointing to delayed detection and access as the main driver of mortality. Third, geography is widening: the 2026 outbreak is the first to be reported in four countries across eastern, central and western Africa, and the first in which imported Bundibugyo cases reached Kenya and Nigeria.

Diagnostics: confirmation is molecular

Because early Ebola is clinically indistinguishable from malaria, typhoid, shigellosis, meningitis and other viral haemorrhagic fevers, laboratory confirmation is essential. The WHO lists four confirmatory methods: reverse-transcriptase polymerase chain reaction (RT-PCR) to detect viral RNA, antibody-capture ELISA, antigen-capture detection tests, and virus isolation by cell culture. In practice, RT-PCR on blood is the standard: it is sensitive, species-specific in design, and fast enough to guide isolation decisions. Antigen-capture tests can be used during active infection, and IgM/IgG serology helps identify recent or past infection but is not the tool for acute diagnosis.

Two diagnostic limitations deserve emphasis. First, assay design is species-sensitive: in the 2007 Bundibugyo outbreak, assays designed for other ebolaviruses initially missed the new virus, and species differences still affect diagnostic target choice today. Any test deployed in a Bundibugyo or Sudan virus outbreak must be validated against that species, not assumed from Zaire-virus data. Second, samples from suspected cases are an extreme biohazard: testing of non-inactivated samples requires maximum containment, and safe transport (triple packaging) is part of the diagnostic chain. Point-of-care molecular platforms run closer to patients have been a genuine advance, but they still depend on supply chains, electricity and trained staff — all fragile in the conflict-affected areas where the 2026 outbreak is spreading.

Vaccines: the evidence, appraised

Ervebo (rVSV-ZEBOV). This is a recombinant, replication-competent vesicular stomatitis virus in which the VSV glycoprotein gene is replaced by Ebola virus GP. It is given as a single intramuscular dose. Its pivotal evidence is the Ebola ça suffit ring vaccination trial in Guinea (2015), a cluster-randomised, open-label trial in which rings of contacts and contacts-of-contacts of confirmed cases were randomised to immediate vaccination or vaccination delayed by 21 days. In the interim analysis (April–July 2015, 90 clusters, 7,651 people), there were no cases with onset 10 days or more after randomisation in the immediate group, versus 16 cases in seven clusters in the delayed group: vaccine efficacy 100% (95% confidence interval 74.7–100%). The final analysis, with 11,841 participants, found no cases 10 days or more after vaccination among 5,837 vaccinees, versus 23 cases among those not vaccinated, again yielding 100% efficacy with a wide lower confidence bound (68.9% in the randomised comparison).

Appraisal: the trial’s design was ingenious — ring vaccination both tested the vaccine and delivered it where it was needed — but the headline “100% effective” needs its caveats. The estimate rests on a small number of cases, hence the wide confidence interval; the trial was open-label; and the 10-day window means cases incubating before vaccination were excluded by design, which is correct for measuring vaccine effect but means the figure is not “nobody vaccinated ever got Ebola”. Effectiveness at ring level (including unvaccinated ring members) was about 75–76%, consistent with a real but partial population effect. None of this undermines the conclusion that Ervebo is highly protective against Zaire ebolavirus; it does mean the figure should be quoted with its interval, and that it says nothing about Bundibugyo or Sudan virus, against which Ervebo has no demonstrated efficacy. Ervebo received FDA approval on 19 December 2019 for adults, and WHO recommends it as part of outbreak response for Ebola virus disease.

A second vaccine regimen, Zabdeno and Mvabea (Janssen), is also approved for preventive use against Ebola virus disease: a two-dose schedule (an adenovirus-vector prime followed by a modified vaccinia Ankara boost). It is a preventive product for at-risk populations, not the product used for ring vaccination during outbreaks, and like Ervebo it targets Ebola virus (Zaire), not Bundibugyo or Sudan virus.

Bundibugyo and Sudan vaccines. There are no licensed vaccines for either species. The most important current study is the Tumaini trial, launched in Bunia, DRC, on 23 October 2026. It is an adaptive, ring-vaccination platform trial: it first tests whether Ervebo offers cross-protection against Bundibugyo virus — animal data suggest it should not be assumed to — and is designed to add two Bundibugyo-specific candidates, an Oxford adenovirus-vector vaccine and a Moderna mRNA vaccine, as they clear earlier-stage testing. Its design is a direct lesson from Ebola ça suffit: testing vaccines inside an outbreak, in the rings around real cases, rather than waiting for the next one.

Treatments: the PALM trial and what it proved

Before 2019 there was no proven specific treatment for Ebola. The PALM trial (Pamoja Tulinde Maisha, “together we save lives”) changed that. Conducted in the DRC during the 2018–2019 outbreak and sponsored jointly by WHO, the US National Institutes of Health and the DRC’s Institut National de Recherche Biomédicale, it randomised 681 patients with RT-PCR-confirmed Ebola virus infection to one of four intravenous therapies, all on top of optimised supportive care (intravenous fluids, daily laboratory testing, correction of blood sugar and electrolytes, and antibiotics and antimalarials as indicated): ZMapp (a three-antibody cocktail, the control), remdesivir (a nucleotide analogue), mAb114 (a single antibody isolated from a 1995 Kikwit survivor, now Ebanga/ansuvimab), or REGN-EB3 (a three-antibody cocktail, now Inmazeb). Randomisation was stratified by viral load (RT-PCR cycle threshold) and treatment site.

An interim analysis led the data and safety monitoring board to stop the ZMapp and remdesivir arms: 28-day mortality was 33.5% with REGN-EB3 (52 of 155) versus 51.3% in the contemporaneous ZMapp subgroup (79 of 154), and 35.1% with mAb114 (61 of 174) versus 49.7% with ZMapp (84 of 169). Remdesivir (53.1%) did not differ significantly from ZMapp. Two findings matter as much as the headline numbers. First, benefit was strongly graded by timing and viral load: patients who arrived within a day of symptom onset died far less often than those who arrived after five days, and mortality fell to around 10% in patients with low viral load who received antibody treatment. Second, even the best arm still lost a third of its patients — these drugs are a major advance, not a cure, and supportive care remains essential.

Appraisal: PALM is the strongest therapeutic evidence in Ebola’s history — randomised, controlled, conducted under outbreak and conflict conditions, with a pre-specified interim analysis and a hard endpoint (death at 28 days). Its limitations are worth stating: it was open-label; the REGN-EB3 arm was added partway through, so its control comparison uses a contemporaneous subgroup rather than the full control arm; and it enrolled only Zaire-virus patients, so it provides no evidence for Bundibugyo or Sudan virus disease, where the antibody epitopes differ. The FDA approved Inmazeb in October 2020 and Ebanga in December 2020 for Zaire ebolavirus infection in adults and children, and WHO has made strong recommendations for either antibody for Ebola virus disease. Inmazeb is a single weight-based intravenous infusion of three antibodies (atoltivimab, maftivimab, odesivimab) binding non-overlapping GP epitopes; Ebanga is a single monoclonal antibody (ansuvimab) binding the GP receptor-binding region. Both work by blocking viral entry; neither has established efficacy against other ebolavirus species, and antibody escape through GP mutation remains a theoretical concern that argues for cocktails and genomic surveillance rather than single antibodies used indefinitely.

Supportive care deserves equal billing in any honest appraisal. The WHO’s “optimised supportive care” package — aggressive fluid and electrolyte management, monitoring and correction of hypoglycaemia, treatment of co-infections, oxygen and organ support — is what every PALM patient received, and it is the only treatment available for Bundibugyo and Sudan virus disease today. In the 2026 outbreak, the single most effective mortality-reducing intervention available right now is not a drug: it is getting patients into treatment centres earlier, which is why community deaths are the outbreak’s most damning statistic.

Infection control and outbreak response: what actually stops Ebola

Ebola outbreaks end through a package, not a single intervention: rapid case identification and isolation, laboratory confirmation, contact tracing with 21-day follow-up, safe and dignified burials, infection prevention and control, community engagement, and — for Zaire ebolavirus — ring vaccination. The 2026 outbreak shows every element under strain: contact follow-up in the DRC ran below the 85% target through September and early October 2026 (80.4% in the latest WHO period, 23,741 of 29,535 contacts reached), about 80% of new cases were recently arising outside known transmission chains, and health workers have been attacked while responding. No vaccine compensates for a broken contact-tracing chain, and no tracing chain works without community trust — the variable that most often decides whether an outbreak lasts months or years.

Research gaps: the honest list

1. Pan-ebolavirus countermeasures. The central gap: vaccines and antibodies that protect against Zaire, Sudan and Bundibugyo viruses, or at least proven cross-protection data. Tumaini is the first test of the cross-protection question in humans.
2. Correlates of protection. There is still no established immune marker that reliably predicts protection against Ebola, which forces every vaccine to be tested in outbreak efficacy trials rather than by immunogenicity alone.
3. Persistence and relapse biology. Why virus persists in immune-privileged sites, what triggers relapse, and whether persistence can seed new outbreaks (as suspected in Guinea in 2021) are only partly understood, and they determine how long survivor programmes must run.
4. Diagnostics for non-Zaire species. Species-validated rapid molecular tests for Bundibugyo and Sudan viruses, deployable without BSL-4 laboratories, are an unmet need that the 2026 outbreak has made urgent.
5. Special populations. Evidence in pregnancy, young children and people with HIV or malnutrition remains thin, and dosing and vaccine eligibility are often extrapolated.
6. Post-Ebola syndrome. The mechanisms of joint, eye and neurocognitive sequelae, and the best models of survivor care, are poorly characterised.
7. Access and delivery. Even proven products fail when cold chains, conflict, funding gaps and health-worker safety break down. The 2026 outbreak is a delivery failure as much as a scientific one, and delivery research (decentralised testing, community surveillance, ring vaccination logistics) deserves the same status as laboratory research.

Conclusion

Ebola’s story since 1976 is one of the clearest arcs in infectious disease: from a mysterious, untreatable haemorrhagic fever to a disease with a proven vaccine, proven antibody treatments and a validated outbreak playbook — for one species. The 2026 Bundibugyo epidemic is the counterweight: species-specific countermeasures leave most of the genus uncovered, and the tools that matter most meanwhile — fast detection, isolation, contact tracing, safe burials and excellent supportive care — depend on health systems and community trust that cannot be improvised during a crisis. The next chapter will be judged by whether it closes the species gap before the next spillover, not after it.

Sources

1. World Health Organization. Ebola disease (fact sheet, updated 24 April 2025). Key facts: average CFR around 50%, range 25–90%; three viruses cause large outbreaks (EBOV, SUDV, BDBV); approved vaccines and treatments only for Ebola virus; diagnostics (RT-PCR, antibody-capture ELISA, antigen-capture tests, virus isolation); two approved vaccines (Ervebo; Zabdeno and Mvabea); WHO strong recommendations for mAb114 or REGN-EB3; persistence in testes, eye, brain, placenta and breast milk; sexual transmission documented up to 15 months after recovery. https://www.who.int/news-room/fact-sheets/detail/ebola-virus-disease
2. World Health Organization. Ebola disease caused by Bundibugyo virus — Democratic Republic of the Congo. Disease Outbreak News, 2026-DON619 (figures as of 6 October 2026; contact follow-up 80.4%; previous Bundibugyo CFR 30% in 2007, 50% in 2012; no approved vaccines or specific treatments for BVD; Kenya imported case; Netherlands medical evacuation). https://www.who.int/emergencies/disease-outbreak-news/item/2026-DON619
3. World Health Organization. Ebola virus disease — Democratic Republic of the Congo. Disease Outbreak News, 2025-DON589 (Kasai 2025: 64 cases, 45 deaths, CFR 70.3%, declared over 1 December 2025). https://www.who.int/emergencies/disease-outbreak-news/item/2025-DON589
4. World Health Organization. Ebola virus disease — Democratic Republic of the Congo. Disease Outbreak News, 2020-DON284 (2018–2020: 3,470 cases, 2,287 deaths, CFR 66%, declared over 25 June 2020). https://www.who.int/emergencies/disease-outbreak-news/item/2020-DON284
5. Kuhn JH, et al. Renaming of genera Ebolavirus and Marburgvirus to Orthoebolavirus and Orthomarburgvirus, respectively, and introduction of binomial species names within family Filoviridae. Archives of Virology 2023 (taxonomy table: six species). https://www.researchgate.net/publication/372887411_Renaming_of_genera_Ebolavirus_and_Marburgvirus_to_Orthoebolavirus_and_Orthomarburgvirus_respectively_and_introduction_of_binomial_species_names_within_family_Filoviridae
6. US Centers for Disease Control and Prevention. Ebola disease outbreak in the Democratic Republic of the Congo and Uganda. Health Alert Network HAN 00530, May 2026 (Ervebo licensed for Orthoebolavirus zairense only; no FDA-approved treatment for Bundibugyo virus disease; not airborne). https://www.cdc.gov/han/php/notices/han00530.html
7. US Centers for Disease Control and Prevention. Ebola outbreak caused by Sudan virus in Uganda. Health Alert Network HAN 00521 (no licensed Sudan virus vaccine; no FDA-approved Sudan treatment; prior Sudan outbreaks about 50% mortality). https://www.cdc.gov/han/php/notices/han00521.html
8. Henao-Restrepo AM, Camacho A, Longini IM, et al. Efficacy and effectiveness of an rVSV-vectored vaccine expressing Ebola surface glycoprotein: interim results from the Guinea ring vaccination cluster-randomised trial. Lancet 2015;386:857–866. DOI: 10.1016/S0140-6736(15)61117-5.
9. Henao-Restrepo AM, Longini IM, Egger M, et al.; Ebola Ça Suffit Trial Consortium. Efficacy and effectiveness of an rVSV-vectored vaccine in preventing Ebola virus disease: final results from the Guinea ring vaccination, open-label, cluster-randomised trial (Ebola Ça Suffit!). Lancet 2017;389:505–518. https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(16)32618-6/fulltext
10. Mulangu S, Dodd LE, Davey RT Jr, et al.; PALM Consortium Study Team. A randomized, controlled trial of Ebola virus disease therapeutics. New England Journal of Medicine 2019;381:2293–2303. (681 patients; 28-day mortality REGN-EB3 33.5%, mAb114 35.1%, ZMapp 49.7%, remdesivir 53.1%; viral load, creatinine and liver enzymes predicted death; symptom duration before admission raised odds of death.) DOI not cited: a DOI lookup for this article failed to resolve during research, so no DOI is given here — verify before publication if a DOI link is required.
11. US National Institutes of Health. Two drugs reduce risk of death from Ebola (PALM trial summary, 2019). https://www.nih.gov/news-events/nih-research-matters/two-drugs-reduce-risk-death-ebola
12. Misasi J, et al.; Carette JE, et al. (NPC1 as intracellular Ebola receptor) and subsequent entry-mechanism literature, summarised in: Ebola virus persistence in immune-privileged niches: molecular mechanisms, host–virus interactions, and therapeutic opportunities. Microorganisms 2026;14:2312. DOI: 10.3390/microorganisms14102312. (Genome ~19 kb, seven proteins NP–VP35–VP40–GP–VP30–VP24–L; macropinocytosis, cathepsin cleavage, NPC1 binding, GP2 fusion; VP35/VP24 interferon antagonism; sGP decoy; persistence sites; 2021 Guinea lineage divergence.)
13. Gong X, et al. Genome-wide CRISPR-Cas9 screen identifies SLC39A9 and PIK3C3 as host factors for Ebola virus entry. PLOS Pathogens 2024. DOI: 10.1371/journal.ppat.1012444.
14. News Agency of Nigeria. WHO says Bundibugyo Ebola spreads to 3 countries, urges stronger regional response (WHO statement as of 20 October 2026: DRC 9,704 confirmed cases, 4,687 deaths, CFR 48.3%, 66 health zones, 8 provinces; Kenya 9 cases, 1 death; Nigeria 1 imported case; 307 of 482 recent deaths in communities). https://nannews.ng/who-says-bundibugyo-ebola-spreads-to-3-countries-urges-stronger-regional-response/
15. Al Jazeera. WHO announces start of Ebola vaccine clinical trial in DR Congo, 24 October 2026 (Tumaini trial, Bunia, 23 October 2026; Ervebo cross-protection test; Oxford adenovirus-vector and Moderna mRNA candidates to join; ring vaccination). https://www.aljazeera.com/news/2026/10/24/who-announces-start-of-ebola-vaccine-clinical-trial-in-dr-congo
16. Al Jazeera. Ebola cases in DR Congo surpass 10,000 in fastest-growing outbreak, 27 October 2026 (DRC government data: 10,060 confirmed cases, 4,887 deaths; ~80% of new cases outside known transmission chains, per DRC reporting). https://www.aljazeera.com/news/2026/10/27/ebola-cases-in-dr-congo-surpass-10000-in-fastest-growing-outbreak
17. United Nations News. ‘Burden and loss’ builds in DR Congo, as Kenya strengthens Ebola containment, October 2026 (Uganda outbreak declared over August 2026; outbreak in eight provinces). https://news.un.org/en/story/2026/10/1168655
18. World Health Organization. Interim advice on sexual transmission of Ebola virus disease (semen testing from 3 months, twice-negative rule, 12-month safer-sex period if untested). Summarised in: Thorson A, et al. Ebola virus: an emerging sexually transmissible infection pathogen. Journal of Virus Eradication / PMC. https://pmc.ncbi.nlm.nih.gov/articles/PMC6111642/

Figures for the ongoing 2026 outbreak are snapshots as of 27 October 2026 and will change; check the WHO Disease Outbreak News page for the latest figures.